A study on employment sustainability among Engineering students using a Statistical and Deep Learning framework

In today's saturated job market, it has been one of the biggest challenges for the fresh engineering graduates from India to secure a satisfactory job offer. In this study, a Statistical and Deep learning (SDL) framework is proposed to predict students' academic achievement in terms of employability. The feature selection part of the framework is carried out with the help of a statistical module. A Deep Learning (DL) module is employed to predict their Cumulative Grade Point Average (CGPA) leading to sustainable employment. The DL module is based on Convolutional Neural Network (CNN) and attention based, stacked Bidirectional Long Short-Term Memory (BiLSTM). Lastly, to improve the interpretability of the framework, an explainability module is incorporated. The findings demonstrate how well the suggested framework forecasts students' employability over the long run and the areas in which work has to be done to make them employable as soon as they graduate. The features based on Binomial distribution yields the lowest Mean Squared Error (MSE) and Mean Absolute Error (MAE) with its superior accuracy over the Gaussian and Poisson distributions.